AI Compute

Twitter

Facebook

Copy Link

Key Facts

Key PlayersNVIDIA, Microsoft, Google, AWS, CoreWeave

Top News

Micron launches HBM4 production for Nvidia Vera Rubin with 2.3x bandwidth boost and 20% efficiency gain

Micron has entered high-volume production of HBM4 36GB 12H memory for Nvidia's Vera Rubin GPU platform, delivering over 2.8 TB/s bandwidth—a 2.3x improvement over HBM3E. Announced at GTC 2026, the company simultaneously began shipping PCIe Gen6 SSDs and SOCAMM2 modules, becoming the first supplier to deliver all three products for the Vera Rubin ecosystem at scale.

Technology3 sources

Micron launches HBM4 production for Nvidia Vera Rubin with 2.3x bandwidth boost and 20% efficiency gain

Technology3 sources

Robot dogs priced up to $300,000 now patrol AI data centers as security demand surges

Data center operators are deploying four-legged robots from Boston Dynamics and Ghost Robotics to secure sprawling AI facilities. With prices ranging from $165,000 to $300,000, these autonomous robot dogs patrol perimeters, detect anomalies, and monitor equipment around the clock, offering a cost-effective alternative to human guards who cost roughly $150,000 annually.

Technology4 sources

Robot dogs priced up to $300,000 now patrol AI data centers as security demand surges

Technology4 sources

UK government invests £45 million in AI supercomputer to accelerate fusion research

The UK is investing £45 million in Sunrise, an AI supercomputer dedicated to fusion research at the UK Atomic Energy Authority's Culham campus. Set to launch in June, the 1.4MW system will model plasma behavior and test nuclear fusion reactor designs virtually before costly physical experiments.

Technology2 sources

UK government invests £45 million in AI supercomputer to accelerate fusion research

Technology2 sources

SK Hynix warns memory shortage could persist until 2030 as AI demand strains wafer supply

SK Group Chairman Chey Tae-won said the global chip wafer shortage will likely last until 2030, far longer than earlier projections. The major DRAM supplier cited artificial intelligence's massive appetite for High Bandwidth Memory as the key driver, with the company expecting more than a 20% shortage of wafers and working on strategies to stabilize DRAM chip prices.

Business3 sources

SK Hynix warns memory shortage could persist until 2030 as AI demand strains wafer supply

Business3 sources

Nvidia unveils BlueField-4 STX to eliminate storage bottlenecks stalling agentic AI inference

Nvidia announced BlueField-4 STX at GTC 2026, a modular AI storage reference architecture that inserts a dedicated context memory layer between GPUs and traditional storage. The platform delivers 5x the token throughput and 4x better energy efficiency compared to CPU-based storage, targeting the key-value cache bottleneck that causes GPU underutilization during extended AI agent sessions.

Technology4 sources

Nvidia unveils BlueField-4 STX to eliminate storage bottlenecks stalling agentic AI inference

Technology4 sources

Nvidia's DGX Station brings trillion-parameter AI models to your desk with GB300 Superchip

Nvidia officially opened orders for the DGX Station, a desktop supercomputer powered by the GB300 Grace Blackwell Ultra Superchip. The system delivers 20 petaflops of compute and 748GB of unified memory, enabling AI professionals to run trillion-parameter models locally without cloud infrastructure. Available from Asus, Dell, Gigabyte, MSI, Supermicro, and HP, the workstation targets researchers and developers building autonomous AI agents.

Technology6 sources

Nvidia's DGX Station brings trillion-parameter AI models to your desk with GB300 Superchip

Technology6 sources

Roche and Nvidia deploy 3,500+ Blackwell GPUs to accelerate AI drug discovery and diagnostics

Swiss drugmaker Roche has deployed over 3,500 Nvidia Blackwell GPUs across the U.S. and Europe, creating the largest pharmaceutical AI infrastructure in the industry. The expansion aims to accelerate drug and diagnostics development while reducing timelines and costs through advanced AI computing capacity.

Technology3 sources

Roche and Nvidia deploy 3,500+ Blackwell GPUs to accelerate AI drug discovery and diagnostics

Technology3 sources

Nvidia unveils Vera Rubin Space Module for orbital AI data centers as space computing race heats up

Nvidia announced the Vera Rubin Space Module at GTC 2026, claiming up to 25 times more AI compute than the H100 for orbital inference workloads. Six commercial space companies have already deployed the platform, while CEO Jensen Huang acknowledged the significant challenge of cooling computer chips in space where there's no air convection.

Technology18 sources

Nvidia unveils Vera Rubin Space Module for orbital AI data centers as space computing race heats up

Technology18 sources

HPE expands AI portfolio with Nvidia Blackwell and Vera Rubin systems for enterprise deployment

Hewlett Packard Enterprise unveiled a major expansion of its AI infrastructure at Nvidia's GTC conference, introducing new systems built on Blackwell and upcoming Vera Rubin GPUs. The company is scaling its Private Cloud AI service to 128 GPUs and launching air-gapped configurations for sovereign deployments. HPE's Alletra Storage MP X10000 became the first object storage platform to achieve Nvidia-Certified Storage validation.

Technology5 sources

HPE expands AI portfolio with Nvidia Blackwell and Vera Rubin systems for enterprise deployment

Technology5 sources

Intel Xeon 6 selected as host CPU for Nvidia DGX Rubin NVL8 systems at GTC 2026

Intel secured a critical win at Nvidia GTC 2026 as its Xeon 6 processor was selected as the host CPU for Nvidia's next-generation DGX Rubin NVL8 AI systems. The partnership extends Intel's x86 dominance in Nvidia's flagship AI servers, with Xeon 6 delivering 2.3x memory bandwidth improvements and up to 8TB system memory to handle increasingly complex inference workloads and agentic AI systems.

Technology4 sources

Intel Xeon 6 selected as host CPU for Nvidia DGX Rubin NVL8 systems at GTC 2026

Technology4 sources

Nvidia unveils Groq 3 LPU to accelerate AI inference with SRAM-powered architecture

Nvidia introduced the Groq 3 LPU at GTC 2026, a specialized chip designed to accelerate AI inference using SRAM memory instead of traditional HBM. The chip stems from Nvidia's $20 billion acquisition of Groq's intellectual property and promises to deliver faster token generation for chatbots and AI agents through 256-chip LPX rack systems paired with Rubin GPUs.

Technology14 sources

Nvidia unveils Groq 3 LPU to accelerate AI inference with SRAM-powered architecture

Technology14 sources

Digital fruit fly brain walks while human neurons play Doom in biological computer experiments

San Francisco's Eon Systems has created a digital fruit fly brain model that controls a virtual body, displaying walking and grooming behaviors. Meanwhile, Australian startup Cortical Labs developed a code-deployable biological computer using 200,000 human brain cells that can play Doom. These parallel experiments represent significant steps in brain simulation and brain-computer interfaces, though questions remain about computational demands and future applications.

Science2 sources

Digital fruit fly brain walks while human neurons play Doom in biological computer experiments

Science2 sources

Frore Systems reaches $1.64B valuation as chip cooling becomes critical for AI infrastructure

Eight-year-old Frore Systems has secured $143 million in Series D funding at a $1.64 billion valuation, achieving unicorn status with its liquid cooling technology for AI chips. Founded by former Qualcomm engineers, the company pivoted from air-cooling phones to liquid cooling systems after Nvidia CEO Jensen Huang suggested they tackle data center thermal challenges. The startup's LiquidJet Nexus technology promises to unlock higher compute density and efficiency across AI infrastructure.

Startups3 sources

Frore Systems reaches $1.64B valuation as chip cooling becomes critical for AI infrastructure

Startups3 sources

BCE invests C$1.7 billion in Saskatchewan AI data center with CoreWeave and Cerebras

Canadian telecom giant BCE is investing C$1.7 billion to build a 300-megawatt AI data center in Regina, Saskatchewan, partnering with CoreWeave and Cerebras. The facility, expected to generate C$12 billion in economic value, will be Canada's largest purpose-built AI data center. Construction begins this spring, with the first phase operational in early 2027, supporting sovereign AI compute and creating over 800 jobs.

Technology5 sources

BCE invests C$1.7 billion in Saskatchewan AI data center with CoreWeave and Cerebras

Technology5 sources

Micron Technology posts record margins, expands AI memory capacity with Taiwan fab acquisition

Micron Technology delivered a blockbuster Q2 with record margins and revenue across DRAM, NAND, and HBM products. The company closed its $1.8 billion acquisition of a Taiwan manufacturing facility and announced plans for a second site, while signing multi-year strategic customer agreements to meet surging AI memory demand. Analysts raised price targets to as high as $600, though some warn margins may have peaked.

Business3 sources

Micron Technology posts record margins, expands AI memory capacity with Taiwan fab acquisition

Business3 sources

Prev 1…464748…85 Next